The present application focuses on testing the determinant spreading and TH1/TH2 hypothesis in the murine IDDM model of NOD mice. The investigator has previously shown that a TH1 T-cell response to GAD is critical for induction of IDDM in NOD mice. He has also shown that the initial response to GAD is followed by responses to an increasing number of beta cell antigens. He hypothesizes that this is the result of determinant spreading, which may be a generalizable feature of T- cell-mediated autoimmune disease. In the present application, he will test the working hypothesis that a first generation of effector cells are triggered by an initiating determinant, either self or mimicking self. This initial response will be a pro-inflammatory TH1 type response, resulting in the release of lymphokines in the target organ that locally operate antigen presentation. This, in turn, results in a second wave of priming the autoimmune T-cells with distinct specificities for autoantigens. The second wave, by definition, is more diverse with respect to antigen recognition, involving both intra- and intermolecular, spreading. The investigator proposes that if the autoantigen-primed T-cells in the second wave are also TH1 cells, further diversification and amplification occurs and pathology progresses. In the absence of spreading, autoimmunity should not progress. Alternatively, the investigator also proposes that if the autoimmune T-cell pool becomes biased toward TH2 cells, the disease process may be halted. This hypothesis will be tested using congenic strains of NOD mice that display distinct phenotypes with respect to beta cell autoimmunity and development of overt diabetes. Aim 1 will be to define the T-cell response to beta cell antigen in murine IDDM. For this, he will use NOD females where both insulitis and diabetes develop, and partial autoimmune NOD mice, NOD.IDD9 mice, where insulitis occurs, but the disease does not progress to diabetes. NOD congenics that do not develop insulitis, strain NOD.IDD3.IDD10, will also be tested. These mice will be evaluated for reactivity to beta cell reactivity and proliferation assays; cytokine-specific ELISA's, and ELISA spot assays, and the kinetics of T-cell responses, and the determinant spreading in TH1 versus TH2 cell type responses, will be determined. Aim 2 will be to test whether initial autoimmunity to GAD triggers a second wave autoimmunity to beta cells. He will test whether GAD- reactive, CD4+ TH1 cells initiate determinant spreading insulitis in diabetes in NOD congenic mice that do not spontaneously develop autoimmunity. He will also compare IDDM induced in mice with a functional immune system, with that in mice without an immune system, in the NOD scid mouse.